Analysis of Foundry defects and Quality improvement of Sand casting

International Research Journal of Engineering and Technology (IRJET)

e-ISSN: 2395-0056

Volume: 09 Issue: 07 | July 2022

p-ISSN: 2395-0072

www.irjet.net

Analysis of Foundry defects and Quality improvement of Sand casting Harsh Verma1, Nischal C N2, Vedant Rajesh Rasal3 1,2,3 Mechanical

Engineering undergraduate, Vellore Institute of Technology, Tamil Nadu, India ---------------------------------------------------------------------***--------------------------------------------------------------------The purpose of this article is to optimize the parameters Abstract - Sand casting is an indispensable manufacturing

of the sand-casting process, including optimal levels. The Taguchi method can be implemented using the following steps:

process which is used widely around the globe. It involves pouring molten metal into moulds having cavities, so as to cast metal into the desired shape of the cavity. Sand casting is a study in itself, and is a blend of fluid mechanics and heat transfer. Common parameters that are varied include- flowing velocity, filling time, porosity of mould material etc. The process of sand casting presents with its own set of challenges, one of them being casting defects. Casting defects compromise the quality of the castings, and make them weaker. The study of casting helps us predict the casting quality and defects without doing actual castings, and hence, save time, money and resources. The main aim behind this is to maximize the sand quality with the help of different methods and also work upon the casting defects therefore improving the overall casting.

1.

To determine the most important variables that influence quality qualities.

2.

Casting defects have been chosen as the most representative quality parameters in the green sandcasting process (cold shut hot tear etc.). The goal of the green sand-casting method is to obtain "reduced casting faults" while reducing the influence of uncontrollable variables.

3.

Make the green sand-casting procedure according to the experimental conditions indicated by the orthogonal array and parameter values you have chosen.

4.

To establish the statistical significance of the parameters, an analysis of variance (ANOVA) table is created. To find the preferred level for each parameter, response graphs are plotted.

5.

Predict the results of each of the parameters at their new optimum values, in addition to the ideal settings of the control parameters.

Key Words: Taguchi Analysis, ANOVA, Orthogonal

arrays, Ishikawa Diagram, Quality Improvement, Sand casting

1.INTRODUCTION Sand casting is an indispensable manufacturing process which is used widely around the globe. It involves pouring molten metal into moulds having cavities, so as to cast metal into the desired shape of the cavity. Sand casting is a study in itself, and is a blend of fluid mechanics and heat transfer.

Common parameters that are varied include- flowing velocity, filling time, porosity of mould material etc. The process of sand casting presents with its own set of challenges, one of them being casting defects. Casting defects compromise the quality of the castings, and make them weaker. The study of casting helps us predict the casting quality and defects without doing actual castings, and hence, save time, money and resources. The main aim behind this is to maximize the sand quality with the help of different methods and also work upon the casting defects therefore improving the overall casting.

Data is collected from the foundries of small scale or medium scale levels. It is collected for a longer period of time and are observed by plotting the graph between the production rate and rejection rate and suitable components are chosen which have higher rejection rates comparatively. It is found that these rejections are mainly due to the sand related defects. Pareto Analysis is used to identify the defects which affects the final casting quality on a larger scale. Based on this testing made, approximately 78% castings were rejected because of cold shut defect and hot tear. Ishikawa Diagram is used to identify the major causes behind the cold shut defect and hot tear. Less fluidity of molten metal, high moisture content in moulding sand, improper gating system, slow and medium pouring, residual stresses in the material and improper mould design were responsible majorly to cause the cold shut defect and hot tear.

2. METHODOLOGY A process which has to be optimized has many process parameters which affects the expected target value directly or indirectly. Optimization involves choosing the right levels of the parameter values to obtain the desired value. This is known as the Static Problem. The noise remains in the whole process but this should not affect the output which is actually the primary aim of Taguchi’s method.

© 2022, IRJET

|

Impact Factor value: 7.529

Three parameters are chosen for experimentation.

1. Pouring temperature

|

ISO 9001:2008 Certified Journal

|

Page 2848

International Research Journal of Engineering and Technology (IRJET)

e-ISSN: 2395-0056

Volume: 09 Issue: 07 | July 2022

p-ISSN: 2395-0072

www.irjet.net

tracked and documented. The lower the amount of casting flaws, the higher the process performance. The objective function to be maximised in this case is:

2. Gating system 3. Pouring time Table -1: First set of Process parameters Sl. No.

Parameter

Level 1

Level 2

1

Pouring Temperature (°C)

1320-1390

1390-1420

2

Gating System

No change

Change

3

Pouring (Second)

15

13

S/N (Signal to Noise) Ratio = -10 log(∑(1/Y2)/n) n = Number of experiments Y = Percentage of approved casting The lesser the S/N value, the better the quality.

Time

S/N ratios are calculated to assess the best experimental process with respect to the levels assigned to each parameter. Signal represents the desired value (Percentage of Approval) and Noise represents undesirable value.

2.3 Selection of Orthogonal array

Two levels are assigned to the parameters with some changes in the values of the parameters (some parameters remain constant for all levels).

The number of control parameters and the interaction of interest influence the choice of an orthogonal array. The number of levels for the control parameters of interest is also a consideration. The standard Taguchi L9 Orthogonal Array (OA) format is chosen from preliminary work with three parameters namely the mould temperature, pouring temperature and as important casting process variables which affect the mechanical properties. The criteria used for choosing the three parameter levels are to explore a maximum range of experimental variables. The assigned levels of parameters are shown in Table 1 and Table 2. The experimental orthogonal array along with their levels are shown in the Table 5.

L4 orthogonal array is used for 3 parameters and 2 levels. It gives us the permutation for the sand casting to be done for various levels.

Process parameters were changed and the same process was followed. The second set of parameters chosen are: 1. Compression Strength 2. Moisture content 3. Permeability

Table – 3: Signal levels

4. Mould Hardness Number Table – 2: Second set of Process parameters Parameter

Range

Level 1

Level 2

Level 3

Compression Strength (g/cm2)

10001300

1000

1260

1300

Moisture Content (%)

3.2-4

3.2

3.5

4

Permeability No

140-190

140

163

190

Mould Hardness Number

85-95

85

90

95

Parameter

Range

Level 1

Level 2

Level 3

Pouring Temperature (°C)

13001500

1300

1320

1350

Permeability (No)

130-180

130

150

180

Sand Particle size (AFS)

50-55

50

52

54

Mould Hardness (Nu)

50-80

50

60

75

Table – 4: Consolidated Process parameters

Three levels are chosen with L9 orthogonal array.

Parameter

2.1 Standard Attributes

Moisture (%) Green (g/cm2)

The quality characteristic to be measured was casting flaws. The most prevalent problems in the foundry were

© 2022, IRJET

|

Impact Factor value: 7.529

|

Level 1

Level 2

Level 3

Content

3.2

3.8

4.3

Strength

1200

1400

1800

ISO 9001:2008 Certified Journal

|

Page 2849

International Research Journal of Engineering and Technology (IRJET)

e-ISSN: 2395-0056

Volume: 09 Issue: 07 | July 2022

p-ISSN: 2395-0072

www.irjet.net

Sand Particle size (AFS)

50

52

54

Mould (nu)

50

60

75

Hardness

Table – 5: Experimental Orthogonal array

Tri al No.

A

B

C

D

E

F

Pouring Temperat ure (°C)

Permeabi lity (No)

Moistu re

Green

Sand Partic le size (AFS)

Mould Hardne ss Numbe r

Conten t (%)

Streng th (g/cm 2)

Figure – 1: S/N Ratios with corresponding levels

1

1300

130

3.5

1200

50

50

2

1300

150

3.5

1500

52

60

3

1300

180

3.5

1700

54

75

4

1320

150

3.5

1200

52

75

Experim ent

Signal Level parameters

5

1320

180

3.5

1500

54

50

Number

Pouring

6

1320

130

3.5

1700

50

60

7

1350

180

3.5

1200

54

60

8

1350

130

3.5

1500

50

75

1

9

1350

150

3.5

1700

52

50

Table – 6: Experimentation Results

The experiments were conducted twice for two same set of parameters. The casting defects that occur in each trial conditions were found and recorded for optimization. It is done with the help of Minitab 16 software by Taguchi Design of Experiment (DOE) technique. Only single defect – Cold shut is considered for the inspection. The average of the casting defects was determined for each trial condition as shown in Table 9. The casting defects are the “lower the better” type of quality characteristics. Lower the better S/N ratios were computed for each of the 9 trials and the values are given in Table 9:

Total Casti ng Pour ed

Numb er of Castin gs appro ved

Percent age of Casting s approv ed

Gati ng

Pouri ng

Syste m

Time

1

1

1

36

27

75%

2

1

2

2

36

33

91.67%

3

2

1

2

36

31

86.11%

4

2

2

1

32

32

100%

Tempera ture

3. EXPERIMENTATION RESULTS

for

Table – 7: S/N Ratios of the experiments performed Experiment Number

Percentage of approved Casting

Signal to Noise Ratio

Mean

1

75%

37.5012

75

2

91.67%

39.2445

91.67

3

86.11%

38.7011

86.11

4

100%

40.0000

100

Table – 8: Optimum parameter values

For example, for trial number 1:

Parameters

Level

S/N

Mean

Optimum Value

2

39.35

93.06

13901420

= -10 log [(42.25+28.09+47.61)/3]

Pouring Temperature (°C)

= -10 log [(117.95)/3]

Gating System

2

39.62

95.84

Change

= -10 log [39.31]

Pouring Time (Second)

2

38.97

88.89

13

η = -10 log [(Ʃy^2i)/3] S/N Ratio = - 10 log [(6.50^2) +(5.30^2) +(6.90^2)/3]

= -15.945

© 2022, IRJET

|

Impact Factor value: 7.529

|

ISO 9001:2008 Certified Journal

|

Page 2850

International Research Journal of Engineering and Technology (IRJET)

e-ISSN: 2395-0056

Volume: 09 Issue: 07 | July 2022

p-ISSN: 2395-0072

www.irjet.net

3.1 Analysis of the results and its effects

Rejection of the final cast products came down to 3.4% which is a good improvement.

The analysis of variance technique (ANOVA) is used to examine the outcomes of the experiments after they have been completed. For several trial settings and parameters that significantly influence casting defects, important factors and/or their interactions are discovered. However, more information is required to arrive at an optimal parameter setting.

 The outcome of conducting the Taguchi Analysis with

the help of Ishikawa Diagram and Pareto Analysis, has been satisfying. This method is efficient in both ways, i.e. when a single defect is identified and Taguchi method is applied to remove or reduce that particular defect and also when this method is used to reduce the defects which are generally seen in the final cast.

Table – 9: Defects percentage and S/N ratios compared with the respective experiment trials Trial No.

% Defects in Experiment

S/N Ratio

Average

1

2

3

Total

1

6.50

5.30

6.90

18.70

-15.9450

6.23

2

5.50

3.80

5.30

14.6

-13.7506

4.87

3

7.34

5.60

7.56

20.5

-16.6884

6.83

4

3.50

4.20

4.01

11.71

-11.8213

3.90

5

6.32

4.90

7.14

18.36

-15.7350

6.12

6

7.43

6.67

6.00

20.1

-16.5215

6.70

7

7.12

5.00

5.75

17.87

-15.5049

5.96

8

3.33

7.32

4.50

15.15

-14.0658

5.05

9

3.50

5.60

1.53

10.63

-11.1501

3.61

 This method not only contributes as one of the efficient and feasible solutions to reduce cast defects directly, but this method has also become an important foundation for many other solutions which were inspired from this technique that were proposed later on too.

REFERENCES [1]

Defects Analysis and Optimisation of Process Parameters using Taguchi DOE Technique for Sand Casting (International Research Journal of Engineering and Technology (IRJET), 2017) by Anil Rathore1, Prof. F. Ujjainwala

[2]

Optimisation of Casting Process Parameters using Taguchi Analysis (International Journal of Mechanical Engineering and Research, 2015) by Saravana Kumar M, Jeya Prakash K

[3]

Reduction of Casting Defects Using Taguchi Method (Journal of Mechanical and Civil Engineering (IOSRJMCE), 2016) by Arun Basil Jacob, Arunkumar O N

[4]

Minimisation of the Casting defects using Taguchi’s Method (International Journal of Engineering Science Invention, 2016) by Harvir Singh, Aman Kumar

[5]

Taguchi Optimisation of Process Parameters on the hardness and Impact energy of Aluminium alloy Sand casting (Leonardo Journal of Sciences, 2013)

by John O Oji, Pamptoks H Sunday

[6]

Optimisation of Green Sand-casting Process Parameters of a Foundry by using Taguchi Method (International Journal of Advanced Manufacturing Technology, 2007) by Sushil Kumar, PS Satsangi, Dr. D R Prajapathi

[7]

Optimization of Sand-Casting Process Parameter Using Taguchi Method in Foundry (International Journal of Engineering Research & Technology (IJERT) by Rasik A Upadhye, Dr. Ishwar P Keswani

[8]

Analysis of Casting Defects in Foundry by Computerised Simulations (CAE) – A New Approach

Figure 2: Graph depicting S/N ratios across the parameters

4. CONCLUSIONS  It was observed that all the first set of parameters provide better results with respect to the cast quality when worked in Level 2. Pouring temperature showed 92-96% of approval, Gating system showed 96% approval and Pouring time of about 84-86% approval.  It is observed that Moisture content and Permeability provide better results at Level 2 whereas, Compression Strength and Mould Hardness Number at Level 3. 20-30% reduction in defects variation could be seen and also the regression value was 85%.

© 2022, IRJET

|

Impact Factor value: 7.529

|

ISO 9001:2008 Certified Journal

|

Page 2851

International Research Journal of Engineering and Technology (IRJET)

e-ISSN: 2395-0056

Volume: 09 Issue: 07 | July 2022

p-ISSN: 2395-0072

www.irjet.net

along with some Industrial Case Studies (IFEX, 2013) [9]

Review on Analysis of Foundry Defects for Quality Improvement of Sand Casting (Int. Journal of Engineering Research and Applications, 2014) by Sunil Chaudhari, Hemant Thakkar

[10]

Review of Optimization aspects for Casting process (International Journal of Science and Research (IJSR), 2015) by Yazad N Doctor, Dr. Bhushan T Patil, Aditya M Darekar

[11]

An Application of Pareto Analysis and Causes Effect diagram for minimization of defects in manual Casting process (International Journal of Mechanical And Production Engineering, 2014) by Aniruddha Joshi, Pritam Kadam

[12]

Casting Defects Analysis in Foundry and Their Remedial Measures with Industrial Case Studies (IOSR Journal of Mechanical and Civil Engineering (IOSR-JMCE)) by Avinash Juriani

[13]

Design Optimization of gating and feeding system through Simulation technique for Sand casting of wear plate (Engineering and Material Sciences) by Sachin L Nimbulkar, Rajendra S Dalu

BIOGRAPHIES Harsh Verma is an undergraduate Mechanical Engineering student at Vellore Institute of Technology, India. He is intrigued by Manufacturing and its Automation.

Nischal C N is an undergraduate student pursuing Mechanical Engineering at Vellore Institute of Technology, India. He is fascinated

by Automobile domain.

Vedant Rasal is an undergraduate student pursuing Mechanical Engineering at Vellore Institute of Technology, India. He is interested in manufacturing domain.

© 2022, IRJET

|

Impact Factor value: 7.529

|

ISO 9001:2008 Certified Journal

|

Page 2852